Header logo is

Contrast discrimination using periodic pulse trains




Understanding contrast transduction is essential for understanding spatial vision. Previous research (Wichmann et al. 1998; Wichmann, 1999; Henning and Wichmann, 1999) has demonstrated the importance of high contrasts to distinguish between alternative models of contrast discrimination. However, the modulation transfer function of the eye imposes large contrast losses on stimuli, particularly for stimuli of high spatial frequency, making high retinal contrasts difficult to obtain using sinusoidal gratings. Standard 2AFC contrast discrimination experiments were conducted using periodic pulse trains as stimuli. Given our Mitsubishi display we achieve stimuli with up to 160% contrast at the fundamental frequency. The shape of the threshold versus (pedestal) contrast (TvC) curve using pulse trains shows the characteristic dipper shape, i.e. contrast discrimination is sometimes “easier” than detection. The rising part of the TvC function has the same slope as that measured for contrast discrimination using sinusoidal gratings of the same frequency as the fundamental. Periodic pulse trains offer the possibility to explore the visual system’s properties using high retinal contrasts. Thus they might prove useful in tasks other than contrast discrimination. Second, at least for high spatial frequencies (8 c/deg) it appears that contrast discrimination using sinusoids and periodic pulse trains results in virtually identical TvC functions, indicating a lack of probability summation. Further implications of these results are discussed.

Author(s): Wichmann, FA. and Henning, GB.
Pages: 74
Year: 2000
Month: February
Day: 0

Department(s): Empirical Inference
Bibtex Type: Poster (poster)

Digital: 0
Event Name: 3. T{\"u}binger Wahrnehmungskonferenz (TWK 2000)
Event Place: T{\"u}bingen, Germany
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: Web


  title = {Contrast discrimination using periodic pulse trains},
  author = {Wichmann, FA. and Henning, GB.},
  pages = {74},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  month = feb,
  year = {2000},
  month_numeric = {2}